Cardiac-derived neutrophil chemotactic factors: Detection in coronary sinus effluents of patients undergoing myocardial revascularization

Cardiac-derived neutrophil chemotactic factors: Detection in coronary sinus effluents of patients undergoing myocardial revascularization Recent studies from our laboratory have demonstrated the release of neutrophil chemotactic factors from isolated rabbit hearts perfused with cardioplegic solutions and from ischemic dog hearts after coronary artery occlusion for 1 hour. On the basis of these animal studies, a test is now made of the hypothesis that neutrophil chemotactic factors are released by myocardial tissues of patients who undergo surgical myocardial revascularization. By means of modified Boyden chambers, the levels of neutrophil chemotactic factors were measured in effluent coUected from the coronary sinuses of six patients undergoing cardiopulmonary bypass during periods of cold cardioplegia. Plain cardioplegic solutions were also analyzed. The standard formyl-methionyl-Ieucyl-phenylalanine, a stimulant of neutrophil recruitment, was used as a positive control solution. Results indicated the recovery of significantly high levels of neutrophil chemotactic factors in patient samples (i.e., 128 % ± 19 % of formyl-methionyl-leucyl-phenylalanine) compared with control plain cardioplegic solution (less than 5 % of formyl-methionyl-Ieucyl-phenylalanine) (p < 0.0001). A standard checkerboard analysis indicated that the observed activity is chemotactic (i.e., directed migration) and not chemokinetic (i.e., random migration~ This study also showed that these factors are proteins of a molecular weight in excess of 300 kd and exhibit in vivo activity by recruiting neutrophils into rabbit skin. The absence of immune ceU-derived chemoattractants such as interleukin-l and leukotriene B4 in these coronary sinus effluents suggests that the observed chemotactic activity is cardiac derived. Results of this investigation therefore demonstrate the release of neutrophil chemotactic factors by ischemic human hearts during cardiopulmonary bypass. (J THORAC CARDIOVASC SURG 1992;103:952-9)

Salwa A. Elgebaly, PhD, Fayyaz H. Hashmi, MD, Stuart L. Houser, MD, Medhat E. Allam, MD, and Kathleen Doyle, RN, CCP, Farmington, Conn.

Acute inflammatory processes are generally initiated and amplified by chemical signals called neutrophil chemotactic factors (NCF).1 These factors recruit neutrophils into sites of infection or damaged tissue, increase cell adhesion, and activate neutrophils to release toxic agents, including oxygen metabolites and proteases.r" A From the Surgical Research Center, Department of Surgery, University of Connecticut School of Medicine, Farmington, Conn. Supported by National Institutes of Health Research Grant HL 42654 and by a gift from Drs. Surendra K. Chawla, Stuart Houser, Fayyaz Hashmi, and Timothy L. Lehmann. Received for publication April 23, 1990. Accepted for publication Feb. 7,1991. Address for reprints: Salwa A. Elgebaly, PhD, Department of Surgery, University of Connecticut Health Center, School of Medicine, Farmington, CT 06032. 12/1/29020

952

number of studies have shown a significant accumulation of neutrophils in infarcted human myocardium' and in animal myocardium made ischemic by coronary artery occlusion.v" The role of neutrophils in postischemic damage has been extensively studied and has been shown to be associated with (I) capillary no-reflow phenomena.? (2) ventricular arrhythrnias.!? and (3) myocardial cell injury. II Factors that recruit neutrophils into the ischemic myocardium may be derived from activated serum protein (complement component C5a),12 activated immune cells (leukotriene B4 [LTB4l and interleukin-l [IL-l D, 13,14 and/or ischemic myocardial tissue itself, as we 15-20 recently demonstrated. The capability of ischemic myocardial tissues to release NCF was shown in a number of species, including isolated bovine, rabbit, and porcine hearts, IS, 16 as well as chicken hearts (unpublished observations). The release of NCF from canine hearts

Volume 103

Number 5 May 1992

was also demonstrated in dogs that had undergone coronary artery occlusion for I hour followed by reperfusion for 2 hours.'? These cardiac-derived factors not only stimulated neutrophil migration but also activated neutrophils to release the digestive enzyme glucosaminadase." Preliminary biochemical characterization revealed that they are proteinaceous in nature, heat unstable, water soluble, inactivated by freeze/thawing, and exhibit a molecular weight in excess of 100 kd. 16-18 Gel filtration high-performance liquid chromatography further confirmed that the high molecular weight cardiac-derived factors are different from C5a (15 kd), IL-I (17 kd), and LTB4 (366 d).18 The release of these mediators from rabbit ischemic hearts was not suppressed by the administration of the antiinflammatory drug ibuprofen'? but was significantly suppressed by the administration of the creatine analog cyclocreatine.P Based on the rabbit and canine studies described, we initiated the present investigation of human hearts. We tested the hypothesis that ischemic myocardial tissues of patients who undergo cardiac operations for myocardial revascularization release NCF. Cardioplegic solutions collected from the coronary sinuses of six patients were assayed for (I) the presence and potency of NCF, (2) the in vivo activity of the recovered factors, (3) preliminary biochemical characterizations of factors recovered in patients' effluent to determine whether these human N CF are proteins of high molecular weights consistent with our previously reported cardiac-derived N CF, and (4) the presence of other low molecular weight chemoattractants such as IL-I and LTB4, which might be released by contaminant blood cells in the cardioplegic solution preparations.

Patients and methods Collection of cardioplegic solutions. Samples were collected from six male patients aged 50 to 68 years (mean 56.7). All patientsunderwent aorta-coronary artery bypass for treatment of symptomatic coronary artery occlusive disease. The mean numberof grafts per patient was 3.3. The operations were performed through median sternotomies.Cardiopulmonary bypass was achieved by ascending aortic and bicaval cannulation. Moderatesystemichypothermia (28° to 30° C) was established in each case. Hearts were arrested with a cold nonoxygenated cardioplegic solution (IL of Plasma-Lyte contained 30 ml of 50% dextrose, 24 mEq of sodium bicarbonate, 20 mEq of potassium chloride, 20 IU of soluble insulin, and 1 mEq of calciumchloride) infused into the aortic root through a 14-gauge angiocatheter with the aorta crossclamped above. An initial dose of cardioplegicsolution of 750 ml was given, followed by bolus doses of 250 ml after completion of distal graft anastomoses. The average volumeof cardioplegicsolutioninfused was 1558 L (I 350 to 2000 L), which maintained myocardial temperaturebelow20° C. The effluentof this infusion was collected byan indwelling coronary sinus catheter placed through the

Cardiac-derived inflammatory mediators

953

right atrium before the start of cardiopulmonary bypass. The catheter was removed when the aorta was unclamped. The average clamp time was 52.2 minutes (45 to 78 minutes). Chemotactic activity was measured in cardioplegic solutions collected at the end of the aortic crossclamping. Approval by the University of Connecticut Institutional Review Board was obtained before the described procedures were conducted. Patient consent forms were also available. Chemotaxis assay. Neutrophil chemotactic activity was determined with modified Boyden chambers as previouslydescribed.16, 20 Formyl-methionyl-leucyl-phenylalanine (f-MLP) (I 0-9M) was the positivecontrol for 100%chemotactic response. Hanks balanced salt solution (HBSS) was the negativecontrol for random migration. Rabbit neutrophils were used as indicator cells. Neutrophils were obtained from the peritoneal cavity of New Zealand White rabbits 4 hours after intraperitoneal injection of 300 ml of NaCl 0.15 mol/L containing 0.1% oyster glycogen (Sigma Chemical Co., St. Louis, Mo.). Neutrophils were adjusted to a final density of 2.5 X 106 cells/rnl in HBSS containing 0.1% bovine serum albumin. A 150,Ill amount of neutrophil suspensionwas placed in the upper compartment of each chamber containing Millipore membrane with a porosityof8 ,uM (Millipore Corporation, Bedford, Mass.). The lower compartment of the chamber contained 150 ,Ill f-MLP, HBSS, plain cardioplegic solutions, or cardioplegicsolutionsobtained from patients undergoing bypass (i.e., undiluted and diluted 1:3 and 1:9 in HBSS). Chambers were then incubated for I hour at 37° C in 5% carbon dioxide/ 95% air atmosphere to permit cell migration across membrane. After incubation, filters were removed, fixed (100% ethanol), stained (IOO% hematoxylin), and mounted on glass slides for counting with an Optomax Image Analyzer (Optomax, Burlington, Mass.) of high-power fields (X40). Cell migration in response to the putative chemoattractants was expressed as percent chemotactic response induced by f-MLP. Chemotactic index (CI) is defined as distance traveled in the filter X cell number. A total of six readings (duplicate filters X three readings/per filter) were taken per each patient sample. Neutrophil chemotactic activity (%f-MLP)

ClSamples - CIHBSS - - - = - - - - - - X 100 C1f-MLP - CIHBSS

All data were expressed as mean value ± standard error of the mean. Student's unpaired t test was used for statistical analysis. A minimum oftwo chemotaxis repeats were performed for each patient sample (total of six patients). To determine whether the observed neutrophil migration in patient samples is due to increased directed migration (i.e., chemotaxis) or increased random migration (i.e., chemokinesis), we performed a standard checkerboard analysis as previously described." Briefly, varying amounts (undiluted and diluted 1:3and 1:9in HBSS) of cardioplegic solutionsobtained from three patients were added to both the upper and lower compartments of the Boydenchamber and incubated at 37° C in a 5% carbon dioxide/95% air atmosphere for 1 hour. Preliminary biochemical characterization. Initial biochemical characterization was performed in patient cardioplegic solutions as previously described.16 Sensitivity to proteases. Aliquots (167 mg gel/25 U protease) of agarose-conjugated type VIII-A subtilisin (Sigma) were washed twice with phosphate-buffered saline (pH 7.2).16 The packed gel preparations were suspended in 500 ,Ill phos-

954

The Journal of Thoracic and Cardiovascular Surgery

Elgebaly et al.

140 >.

:= ~ ()

Cl:

120 100

~-

80

~~

60

s~

i~

40 20 0

Plain

Patient

Cardlopleglc Solutions

Fig. 1. Neutrophil chemotactic activity recovered in effluents collected from coronary sinuses of six patients who underwent surgical myocardial revascularization. Plain cardioplegic solution was also tested. FMLP, Formylmethionyl-Ieucyl-phenylalanine.

phate-buffered saline, and lOO,u1 volume was mixed in 800,u1 patient cardioplegic solutions. The suspensions were incubated at 37° C for 90 minutes and then centrifuged at 5000 rpm for 5 minutes. The supernatant solutions were collected and assayed separately for chemotactic activity. The advantage of this approach (i.e., removal of agarbse-linked proteases) is that it eliminates the interferences of proteases with neutrophil migration in the Boyden chambers. Ultrafiltration with Amicon membranes. Patient cardioplegic solutions (10m!) were fractionated with Amicon membranes of molecular weight cutoff levels of 300 kd (Amicon Div./W. R. Grace & Co., Beverly, Mass.)." Fractions collected above the filters (I ml) were assayed for chemotactic activity as undiluted, I:3, and 1:9 in HBSS. In vivo activity. Patient cardioplegic solutions (IO m!) were concentrated tenfold (i.e., to I ml) with an Amicon membrane of 300 kd. Approximately 200,u1of the fluid retained by the filter was injected intradermally into shaved rabbit back skin. Control solution involved injection of f-MLP (10- 9 M) and HBSS. After 24 hours, samples of skin were removed and processed for analysis with a light microscope to determine neutrophil accumulation. Levels of LTB4 and IL-l in the coronary sinus effluents. The levels of LTB 4 were measured in the patient effluents with a commercially available radioimmune assay kit (Du Pont Company, Wilmington, Del.). A standard curve was prepared by diluting the stock (40 ng/rnl LTB4) with assay buffer to achieve concentrations ranging from 5 ng/rnl to 0.125 ng/mI. 2 l • 22 A volume of 100,u1sample of each patient effluent was mixed in duplicate with 100 ,Ill assay buffer, 100 ,Ill tracer solution eH-LTB4), and lOO,u1 antiserum. The mixture was incubated overnight (I8 hours) at 4° C. At the end of the overnight incubation, tubes were placed in an ice bath, and 500 ,Ill charcoal suspension was added to each tube. Tubes were then placed individually in a vortex for 2 to 5 minutes, incubated in the ice bath for IS minutes, and centrifuged in a refrigerated centrifuge at 1000 to 2000 g for IS minutes. Supernatant solutions were then removed and placed into scintillation vials con-

taining scintillation cocktail. Radioactivity was measured with a liquid scintillation counter. Nonograms of LTB4 present in each patient sample were calculated by interpolation from the standard curve. The levelsof IL-I were measured in the patient effluents with a commercially available enzyme-linked immunospecific assay (Cistron Biotechnology, Inc., Pine Brook, N.J.). A standard curve was prepared by diluting the stock (50 ng/rnl IL-l) with phosphate buffer to achieve concentrations ranging from 20 pg/ml to 1000 pg/ml. 22 A 100 ,Ill volume of each standard, control, and patient sample (in duplicate) was added into a 96-well plate containing polyclonal rabbit anti-IL-l. The wells were covered, and the plate was incubated at 37° C for 2 hours. The wells were then washed three times with the wash buffer, and 100 ,Ill of IL-I antiserum was added to each well. Wells were covered and the plate reincubated for an additional 2 hours at 37° C. Wells were washed three times with the wash buffer, and 100 ,Ill of antirabbit IgG conjugated to horseradish peroxidase enzyme was added to each well. Wells were covered, and the plate was incubated at room temperature for 30 minutes and then washed. A 100 ,Ill volume of the enzyme substrate was added to all wells, and the plate was incubated at room temperature for IS minutes. At the end of the incubation time, 50 ,Ill of 4N sulfuric acid was added to each well. The plate was read at a 490 nm wavelength with a standard enzyme-linked immunospecific assay reader. The amount of IL-I present in each patient sample was determined by interpolation from the standard curve. Results Recovery of neutrophil chemotactic activity in cardioplegic solutions obtained from patients undergoing cardiac bypass operations. Samples collected from six patients undergoing cardiac bypass operations were assayed individually for the presence and potency of NCF. Chemotactic activity as a percent f-MLP was cal-

Volume 103 Number 5 May 1992

Cardiac-derived inflammatory mediators

955

3000-,----------------------------, ~

:E (j

ce,.

.2~ -

c

~ :; 2000 I-MLP 0;:

~~

.J:. 0

ue _ Gl

-au -.J:.

g::I

1000

Gl

Z

HBSS 500 '-------.-------....---------r---~ 0.11 0.33 1.0

Dilutions Fig. 2. Dose-response curveof neutrophil chemotactic activity recovered in coronary sinuseffluent of a patient who underwent surgical myocardial revascularization. Activity was tested as undiluted (1.0) and diluted to 1:3 and 1:9 in HBSS. culated at the three tested dilutions (i.e., undiluted, 1:3, and 1:9). The dilution that showed the highest activity for each patient was used in our calculations. In general, samples diluted 1:3 and 1:9 showed the highest response. As can be seen in Fig. I, significant levels of NCF (128 ± 9 of f-MLP) were recovered in the six examined patients. Plain cardioplegic solutions, on the other hand, showed levels equal to random migration induced by HBSS (i.e., less than 5%) (p < 0.00001). Fig. 2 presents a dose-response curve of chemotactic activity recovered in the cardioplegic solution of one patient who underwent myocardial revascularization. In this patient the highest induced neutrophil migration calculated as chemotactic index was again seen at the 1:9 dilution followed by 1:3 dilution. Checkerboard analysis. Checkerboard analysis (Table I, representative of three patients) was performed on a cardioplegic solution of a patient who underwent myocardial revascularization. The left column of the table presents a similar dose-response to that seen in Fig. 2, in which the highest chemotactic response was seen when the cardioplegic sample was diluted 1:9 in HBSS. In this dilution, the migration index increases to 8.1-fold, which is significantly greater than the random migration index (i.e., value of one). The chemokinetic index seen along the diagonal axis shows decreased migration as the net concentration of supernatant in the upper and lower chambers increases. This index was maximal at random migration (i.e., 1.0) when HBSS was placed in the upper and lower chambers. When values below the diagonal werecompared with those above the diagonal, it was clear that migration was greater when the concentrations of

cardioplegic solutions were greater in the lower chamber than the upper chamber, indicating that the activity is chemotactic. Biochemical characterization. The sensitivity of NCF to protease treatment was tested. A reduction in activity of 58% ± 5% was observed after treatment with the bacterial protease subtilisin for 90 minutes, which suggests that the factors are protein in nature. Ultrafiltration with the Amicon membrane showed that most of the activity (149% ± 16% f-MLP) was retained above the filter. This suggests that the factors exhibit a molecular weight of greater than 300 kd. In vivo activity. Materials retained above a 300 kd filter were injected intradermally into rabbit back skin. After 24 hours samples were removed and processed to be studied under a light microscope. As can be seen in Fig. 3, significant neutrophil accumulation to the epidermal layer was evident. This response was similar to that induced by f-MLP. HBSS did not induce neutrophil accumulation. Levels of LTB 4 and IL-l. No detectable levels of IL-I were measured in patients' coronary sinus effluents. Similarly, the levels of LTB4 were below the normal serum levels (i.e., 1.5 ng/ml)21 in all patient samples, which indicates that the observed chemotactic activity recovered in the patients' effluents is not due to IL-I or LTB4. Discussion The objective of this study was to determine the presence of inflammatory mediators such as NCFs in effluent collected from the coronary sinuses of patients undergoing surgical myocardial revascularization. High levels

956

The Journal of Thoracic and Cardiovascular Surgery

Elgebaly et al.

Table I. Checkerboard analysis of neutrophil chemotactic activity recovered in cardioplegic solution of a patient who underwent myocardial revascularization Upper Chamber

0

1:9

0

t

..

.c E

1:3

Undiluted

280 t 93

428 t 102

(0.77)

(1.1) 32 t 8 (0.08)

1:9

~

o

... QI

~

.... 0

1768 t 184

1:3

(4.8) Undiluted

"Migration Index

1616 t 89

51 t 20

(4.5)

(0.14)

Chemotactic Index lor I-MLP

= 2243

, 333 Chemotactic Index lor HBSS

of these mediators have been detected in the coronary sinus effluents obtained from six surgical patients. These factors were absent in plain cardioplegic solutions. Similar to factors previously recovered from ischemic rabbit and canine hearts, 15-20 the human factors are proteins of high molecular weights that exhibit in vivo activity by recruiting neutrophils into rabbit skin. On the basis of molecular weight analysis, it is suggested that the cardiac-derived chemotactic factors are different from other known chemoattractants derived from activated immune cells (e.g., IL-l [17 kd] and LTB 4 [366 d]) or serum proteins (e.g., complement component C5a [15 kd]). Furthermore, the absence of IL-l and LTB 4 in the patients' effluent preparations and the absence of C5a in patient plasma after cardiac operations-' exclude a role for these chemoattractants and suggest that the heart is the source of the recovered chemotactic activity. It is of interest to note that although C5a was not reported to be elevated after operation, the levels of other complement components such as C4 and the complement split product C3d were shown to be increased after cardiac operation.i" Neither C4 nor C3d exhibits chemotactic activity. Elevation of the levels of the weak chemoattractant C3a, on the other hand, was reported in one study25 and denied in another.P A study by Van Velzen-Blad and coworkers-" reported normal levels of the complement C3 before and after cardiac operation, whereas the study by Hayase and coworkers-? reported a significant drop of the serum C3 level during cardiopulmonary bypass. Although current myocardial preservation techniques have successfully protected the heart from ischemic injury, the heart remained ischemic during the operation. Studies by Ts'ao and colleaguesf have demonstrated the accumulation of neutrophils into patient myocardium after cardiac operations. Since some of the accumulated

= 516

, 64

neutrophils were found to be degranulated, the authors suggested a role for these inflammatory cells in some of the cardiac complications associated with operations. Similarly, studies by Antonsen.i" Rommelsheim.i? and their colleagues have demonstrated a significant increase in plasma concentrations of the proteolytic enzyme elastase (a product of neutrophil degranulation) during cardiopulmonary bypass. These studies not only indicate neutrophil accumulation into the heart after cardiac operations, but they also demonstrate their activation to release potent digestive enzymes that are known to cause significant cell injury." As reported here, we have detected high levels of NCF in effluents collected from the coronary sinuses of patients who had undergone myocardial revascularization. Elimination of the contribution of immune cell-derived (i.e., IL-l and LTB 4 ) and serumderived (i.e., C5a) chemoattractants suggests that the observed chemotactic activity recovered in patients' effluents is released from the heart itself. The capability of local tissues to release N CF in response to injury has been established recently by our laboratory. These tissues include cornea,30-33 conjunctiva.F stomach." coronary artery,35 vein grafts-" and urinary bladders.l? Neutrophil accumulation has been shown experimentally to occur within the first I to 5 hours in myocardium made acutely ischemic by coronary artery occlusion.v" The role of neutrophils in postischemic damage to the heart has been the subject of many investigations. Neutrophil infiltration has been associated with (I) capillary no-reflow phenomena." (2) ventricular arrhythmias, '0 and (3) myocardial cell injury" by releasing destructive enzymes such as collagenase, elastase, and cathepsin," as well as oxygen radicals such as hydrogen peroxide, superoxide anion, and hydroxyl radicals.' Rowe, Eaton, and Hess 38 showed that canine neutrophils activated in

Volume 103 Number 5 May 1992

Cardiac-derived inflammatory mediators

9 57

Fig. 3. Neutrophilaccumulation 24 hoursinto the epidermal layer after injection of a concentratedpreparationof patients'coronary sinuseffluents (200 ~I of materials retained abovea 300 kd filter). Sampleswere injectedintradermally.

vivo with tetradecanoylphorbol acetate induced significant cardiovascular dysfunction. A number of agents that reduce neutrophil infiltration have, therefore, been tested for their effects on modulatingmyocardial injury. Ibuprofen.v"" BW755C,42, 43 and dipyridamole'f have been shown to suppress neutrophil accumulation in inflammatory lesions and to exert cardioprotective effects. Studies of neutrophil depletion by filtration.f specificantineutrophil antiserum.t'' or administration of hydroxyurea'" have confirmed that myocardial damage is reduced when neutrophils are prevented from invading the ischemic myocardium. Complement depletion with cobra venom factor attenuated neutrophil infiltration into the heart and reduced the area of damage.48,49 Apoprotein (serum protease inhibitor)50 and superoxide dismutase (superoxide anion scavengerj" reduced infarct size by preventing the release of chemotactic factors (i.e., C5a and superoxide-dependent chemotactic factors) and the accumulation of neutrophils at the siteof a potential inflammatory lesion. Both agents might additionally reduce ischemic damage by preventing direct tissue injury induced by superoxide anion and proteases releasedfrom activated neutrophils. These studies provide evidence for the active participation of neutrophils in exacerbating myocardial cell injury after ischemia and infarction. Future understanding of the mechanisms involved in cellular synthesis and release of the newly identified cardiac-derived inflammatory mediators might lead to the developmentof nonsteroidal therapeutic agents that specifically reduce cardiac inflammation after ischemia

without subjecting the host to immunosuppression. Since the continued arrival of neutrophils leads to additional ischemia by plugging capillaries and reducing blood flow to the tissues,52 developing blockers for the cardiac mediators not only will reduce inflammation but also will protect healthy nonischemic areas. Controlling neutrophil accumulation in ischemic myocardium will be useful specifically in a therapeutic approach to patients with acute myocardial infarction- as well as to patients having cardiac operations. In summary, this report documents the recovery of NCF in cardioplegic solutions collected from the coronary sinuses of patients undergoing operations for myocardial revascularization. This study also suggests a mechanism by which myocardial tissues direct the recruitment of neutrophils. As this mechanism becomes more clearly defined, application of this information to the treatment of patients with myocardial infarction or complications of perioperative myocardial ischemia can be addressed. We extend our thanks to Dr. Ayman Nabawi, Ms. Lili Aramli,and Mr. DinoMianofor their technicalsupport,and to Mrs. Joanne D'Aprile for the preparation of the manuscript. REFERENCES 1. Wilkinson P. Chemotaxis and inflammation. 2nd ed. Lon-

don: Churchill Livingstone, 1982. 2. Charo IF, Yuen C, Perez HD, Goldstein 1M.Chemotactic peptides modulateadherenceof human polymorphonuclear leukocytes to monolayers of cultured endothelial cells. J Immunol 1987; 1-36:3412-9.

958

The Journal of Thoracic and Cardiovascular Surgery

Eigebaly et al.

3. Fantone C, Ward PA. Role of oxygen-derived free radicals and metabolites in leukocyte-dependent inflammatory reactions [Review]. Am J PathoI1982;107:397-418. 4. Becker EL, Henson PM. In vitro studies of immunologically induced secretions of mediators from cells and related phenomena. Adv ImmunoI1973;17:94-103. 5. Jackson DM, Millar AM, Nicoll JJ, Connell M, Muir AL. The acute inflammatory response to myocardial infarction: imaging with indium-III labeled autologous neutrophils. Br Heart J 1987;57:23-7. 6. Sommers HM, Jennings RB. Experimental acute myocardial infarction: histologic and histochemical studies of early myocardial infarcts induced by temporary or permanent occlusion of a coronary artery. Lab Invest 1964;13:14911503. 7. Lucchesi BR, Mullane KM. Leukocytes and ischemia-induced myocardial injury. Annu Rev Pharmacol Toxicol 1986;26:201-24. 8. Engler R. Consequences of activation and adenosine-mediated inhibition of granulocytes during myocardial ischemia. Fed Proc 1987;46:2407-12. 9. Schmid-Schonbein GW. Capillary plugging by granulocytes and the no-reflow phenomena in the microcirculation. Fed Proc 1985;46:2397-401. 10. Maisel AS, Gilpin FA, Lewinter M, Henning H, Ross JJ, Engler R. Initial leukocyte count during acute myocardial infarction independently predicts early ventricular fibrillation. Circulation 1985;72(Pt 2):III414. 11. Ramson JL, Jolly SR, Lucchesi BR. Protection of ischemic myocardium by pharmacologic manipulation of leukocyte function. Cardiovasc Rep 1984;5:690-709. 12. Fernandez HN, Hensen PM, Otani A, Hugli TE. Chemotactic response to human C3a and C5a anaphylatoxins. I. Evaluation of C3a and C5a leukotaxis in vitro and under simulated in vivo conditions. J Immunol 1978;120:102-8. 13. Ford-Hutchinson A W, Bray MA, Doig MU, Shipley ME, Smith MJH. Leukotriene B, potent chemokinetic and aggregating substance released from polymorphonuclear leukocyte. Nature 1980;286:264-5. 14. Cybulsky MI, Colditz IG, Movat H. The role of interleukin-I in neutrophil leukocyte emigration induced by endotoxin. Am J Pathol 1986;124:367-72. 15. Elgebaly SA, Kozol R, Bohr M. Release of neutrophil chemotactic factors from ischemic myocardial tissue. Surg Forum 1987;38:276-8. 16. Elgebaly SA, Masetti P, Allam ME, Forouhar F, Bohr M. Cardiac derived neutrophil chemotactic factors: preliminary biochemical characterization. J Mol Cell Cardiol 1989;21:585-93. 17. Elgebaly SA, Allam ME, Houser S, Hashmi F, Forouhar F. Cardiac-derived neutrophil chemotactic factors [Abstract]. J Leukocyte Bioi 1990;45:22. 18. Elgebaly SA, Cohen A, Allam ME, Yekta N. Cardiac derived inflammatory mediators. J Mol Cell Cardiol 1989; 21(suppl):171. 19. Elgebaly SA, Masetti P, Forouhar F, Rossomando EF, Cohen A. Ibuprofen fails to inhibit the release of tissue

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

derived neutrophil chemotactic factors "Nourin" from ischemic hearts. XXVI World Congress of the International College of Surgeons 1988;27:33-6. Elgebaly SA, Allam ME, Rossomando EF, et al. Cyclocreatine inhibits the production of neutrophil chemotactic factors from isolated hearts. Am J PathoI1990;137:l233-41. Klausner JM, Paterson I, Kobzik L, Valeri CR, Shepro D, Hechtman HB. Leukotriene (L T) B4 mediates ischemicinduced respiratory failure. FASEB J 1988;2:A489. Elgebaly SA, Donshik PC, Rahhal F, Williams W. Neutrophil chemotactic factors in tears of giant papillary conjunctivitis patients. Invest Ophthalmol Vis Sci 1991; 32:204-9. Hammerschmidt DE, Stroncek DF, Bowers TK, et al. Complement activation and neutropenia occurring during cardiopulmonary bypass. J THORAC CARDIOVASC SURG 1981;81:370-7. Antonsen S, Brandslund I, Clemensen S, Sifeldt S, Madsen T, Alstrup P. Neutrophil lysosomal enzyme release and complement activation during cardiopulmonary bypass. Scand J Thorac Cardiovasc Surg 1987;21:47-52. Quiroga MM, Miyagishima R, Haendschen LC, Glovsky M, Martin BA, Hogg JC. The effect of body temperature on leukocyte kinetics during cardiopulmonary bypass. J THORAC CARDIOVASC SURG 1985;90:91-6. Van Velzen-Blad H, Dijkstra YJ, Schurink GA,etaI.Cardiopulmonary bypass and host defense functions in human beings. I. Serum levels and role of immunoglobulins and complement in phagocytosis. Ann Thorac Surg 1985; 39:207-11. Hyase S, Shimizu K, Abe T, et al. Effects of open heart surgery with cardiopulmonary bypass on the immune host defense mechanism. Nippon Kyobu Geka Gakkai Zasshi 1982;30:52-65. Ts'ao C, Lin CY, Glagov S, Replogle RL. Disseminated leukocyte injury during open heart surgery. Arch Pathol 1973;95:357-65. Rommelsheim K, Schlebusch H, Orellano L, Schwidetzky F, Klehr U. Polymorphonuclear neutrophils in extracorporeal circulation procedures. Anaesth Intensivther Notfallmed 1987;22:3-7. Elgebaly SA, Herkert N, O'Rourke J, Kreutzer DL. Characterization of neutrophil and monocyte-specific chemotactic factors derived from isolated corneas in response to hydrogen peroxide injury. Am J Pathol 1987; 126:40-50. Elgebaly SA, Downes RT, Bohr M, Forouhar F, O'Rourke J, Kreutzer DL. Inflammatory mediators in alkali-burned corneas: preliminary characterization. Curr Eye Res 1987; 6:1263-74. Elgebaly SA, Miano D, Kreutzer DL, Fishman JB. Cornea derived neutrophil chemotactic factors: intracellular synthesis. Curr Eye Res 1990;9:839-45. Riley MV, Elgebaly SA. The release of a neutrophil chemotactic factor from UV-B irradiated rabbit corneas in vitro. Curr Eye Res 1990;9:677-82. Kozol RA, Downes R, Elgebaly SA. Gastric derived neu-

Volume 103 Number 5 May 1992

trophil chemotactic factors: preliminary biochemical characterization. Dig Dis Sci 1989;34:681-7. 35. Elgebaly SA, Masetti P, Bohr M. The release of neutrophil chemotactic activity from ischemic coronary artery [Abstract]. J Leukocyte Bioi 1987;42(suppl):197. 36. Ruby ST, Allam ME, Gallo MA, Berth S, Elgebaly SA. Release of neutrophil chemotactic factors by canine veins during preparation for arterial bypass. Arch Surg 1990; 125:481. 37. Walzak MP, Elgebaly SA, Allam ME, Oselinsky D. Leukocyte chemotactic factors in interstitial cystitis-a possible biological marker of the disease. Semin Urol 1991;9:131-5. 38. Rowe GT, Eaton LR, Hess ML. Neutrophil-derived free radical-mediated cardiovascular dysfunction. J Mol Cell CardioI1984;16:1075-9. 39. Romson JL, Bush LR, Haack DW, Lucchesi BR. The beneficial effects of oral ibuprofen on coronary artery thrombosis and myocardial ischemia in the conscious dog. J Pharmacol Exp Ther 1980;215:271-8. 40. Jugdutt BI, Hutchins GM, Bulkley BH, Becker LC. Salvage of ischemic myocardium by ibuprofen during infarction in the conscious dog. Am J CardioI1980;46:74-82. 41. Romson JL, Bush LR, Jolly SR, Lucchesi BR. Cardioprotective effects of ibuprofen in experimental regional and global myocardial ischemia. J Cardiovasc Pharmacol1982; 4:187-96. 42. Mullane KM, Moncada S. The salvage of ischemic myocardium by BW755C in anesthetized dogs. Prostaglandins 1982;24:255-6. 43. Jolly SR, Lucchesi BR. Effect of BW755C in an occlusionreperfusion model of ischemic myocardial injury. Am Heart J 1983;106:8-13. 44. Teoh KH, Christakis GT, Weisel RD, et al. Dipyridamole reduced myocardial platelet and leukocyte deposition fol-

Cardiac-derived inflammatory mediators

45.

46.

47.

48.

49.

50.

51.

52.

9 59

lowingischemia and cardioplegia. J Surg Res 1987;42:64252. Engler RL, Dahlgren MD, Morris D, Peterson MA, Schmid-Schoenbein G. Role of leukocytes in the response to acute myocardial ischemia in dogs. Am J Physiol 1986;251:H314-23. Romson JL, Hook BG, Kunkel S, Lucchesi BR. Reduction of the extent of ischemic myocardial injury by neutrophil depletion in the dog. Circulation 1983;67:1016-23. Mullane KM, Read N, Salmon JA, Moncada S. Role of leukocytes in acute myocardial infarction in anesthetized dogs: relationship to salvage by anti-inflammatory drugs. J Pharmacol Exp Ther 1984;228:510-55. Harmann JR, Robinson JA, Gunner RM. Chemotactic activity in the coronary sinus after experimental myocardial infarction: effects of pharmacologic interventions on ischemic injury. Am J Cardiol 1977;40:550-5. Maroko PR, Carpenter CB, Chiarello M, Fishbein MC, Radvany P. Reduction by cobra venom factor of myocardial necrosis after coronary artery occlusion. J Clin Invest 1978;61:661-70. Diaz PE, Fishbein MC, Davis MA, Askenazi J, Maroko PRo Effect of the kallikrein inhibitor aprotinin on myocardial ischemic injury after coronary occlusion in the dog. Am J Cardiol 1977;40:541-9. McCord JM, Wong K, Stokes SH, Petrone WF, English DK. A mechanism for the anti-inflammatory activity of superoxide dismutase. In: Autor AP, ed. Pathology of oxygen. New York: Academic Press, 1982:75-83. Craddock PR, Hammerschmidt DE, Moldow CF, Yamada 0, Jacob HS. Granulocyte aggregation as a manifestation of membrane interactions with complement: possible role in leukocyte margination, microvascular occlusion, and endothelial damage. Semin Hematol 1979;16:140-7.